Information determination method, device and equipment and computer readable storage medium

Abstract

The embodiment of the invention discloses an information determination method, device and equipment and a computer readable storage medium, and the method comprises the steps: obtaining an initial computing power resource parameter, an initial carbon emission parameter and an initial cost parameter of each computing power node, sorting the plurality of computing power nodes based on the initial computing power resource parameter, the initial carbon emission parameter and the initial cost parameter; receiving a computing power request which is sent by the client and carries a computing task and a task label of the computing task; based on the computing power request, a target computing power resource parameter, a target carbon emission parameter, a target cost parameter and a task label of a computing power node corresponding to the task, determining a target computing power node from the sorted computing power nodes; and sending the identifier of the target computing power node to the client to enable the client to send the computing task to the target computing power node based on the identifier of the target computing power node, so that the determined target computing power node meets the requirements of energy conservation and emission reduction, and the computing accuracy of different computing tasks is improved.

Description

Technical Field

[0001] This application relates to the field of communication technologies, and in particular, to a method, apparatus, device, and computer-readable storage medium for information determination. Background Art

[0002] Currently, in a computing power network, when a scheduling center receives a computing power request, it can dispatch the computing power task in the computing power request to a designated central computing power resource or edge computing power resource based on a computing power resource scheduling algorithm; generally, traditional computing power resource scheduling algorithms only consider the impact of the computing power capacity, network topology, and latency parameters of computing power resources on computing power scheduling, making the computing power resource scheduling algorithm unable to meet the requirements of green, low-carbon, circular development and low cost. Summary of the Invention

[0003] To solve the above technical problems, embodiments of this application are expected to provide a method, apparatus, device, and computer-readable storage medium for information determination, which can solve the problem that the computing power resource scheduling algorithm in related technologies cannot meet the requirements of green, low-carbon, circular development and low cost.

[0004] The technical solution of this application is implemented as follows:

[0005] A method for information determination, the method includes:

[0006] Obtain the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters of each computing power node, and sort multiple computing power nodes based on the initial computing power resource parameters, the initial carbon emission parameters, and the initial cost parameters;

[0007] Receive a computing power request sent by a client, which carries a computing task and a task label of the computing task;

[0008] Based on the computing power request, the target computing power resource parameters, target carbon emission parameters, target cost parameters of the computing power node corresponding to the computing task, and the task label, determine a target computing power node from the sorted computing power nodes;

[0009] Send the identifier of the target computing power node to the client, so that the client sends the computing task to the target computing power node based on the identifier of the target computing power node.

[0010] In the above solution, the sorting of multiple computing power nodes based on the initial computing power resource parameters, the initial carbon emission parameters, and the initial cost parameters includes:

[0011] Determine a first weight of the initial computing power resource parameters, a second weight of the initial carbon emission parameters, and a third weight of the initial cost parameters of the multiple computing power nodes;

[0012] For each computing power node, based on the initial computing power resource parameters, the initial carbon emission parameters, the initial cost parameters, the first weight, the second weight, and the third weight of the computing power node, determine the first computing power value of each computing power node; wherein, the first computing power value is used to represent the priority of the computing power node.

[0013] Sort the multiple computing power nodes based on the first computing power value.

[0014] In the above solution, determining the target computing power node from the sorted computing power nodes based on the computing power request, the target computing power resource parameters, the target carbon emission parameters, the target cost parameters of the computing power node corresponding to the computing task, and the task label includes:

[0015] Group the multiple sub-computing tasks based on the computing power request and the task label to obtain multiple groups of sub-computing tasks; wherein, the computing task includes the multiple sub-computing tasks; the levels of each group of sub-computing tasks are different.

[0016] For each group of sub-computing tasks, determine the first target weight of the target computing power resource parameters, the second target weight of the target carbon emission parameters, and the third target weight of the target cost parameters.

[0017] Determine the target delay parameter of the computing power node corresponding to each group of sub-computing tasks, the target carbon emission parameter of the computing power node corresponding to each group of sub-computing tasks, and the target cost parameter of the computing power node corresponding to each group of sub-computing tasks; wherein, the target computing power resource parameters include the target delay parameter.

[0018] Based on the first target weight, the second target weight, the third target weight, the target delay parameter, the target carbon emission parameter, and the target cost parameter, determine the target computing power node of each group of sub-computing tasks from the sorted computing power nodes.

[0019] In the above solution, grouping the multiple sub-computing tasks based on the task label to obtain multiple groups of sub-computing tasks includes:

[0020] Based on each task label, determine the level of each sub-computing task.

[0021] Determine the sub-computing tasks with the same level as a group of sub-computing tasks.

[0022] In the above solution, determining the target delay parameter of the computing power node corresponding to each group of sub-computing tasks, the target carbon emission parameter of the computing power node corresponding to each group of sub-computing tasks, and the target cost parameter of the computing power node corresponding to each group of sub-computing tasks includes:

[0023] For each group of sub-computation tasks, based on the transmission delay parameter of data transmission between the sub-computation tasks and the computing power nodes and the computing delay parameter of computing the sub-computation tasks on the computing power nodes, determine the target delay parameter;

[0024] Based on the first energy consumption parameter generated by the data transmission, the second energy consumption parameter generated by computing the sub-computation tasks on the computing power nodes, the current carbon emission parameter of the computing power nodes, the first target conversion coefficient, and the second target conversion coefficient, determine the target carbon emission parameter;

[0025] Based on the first energy consumption parameter, the second energy consumption parameter, the current cost parameter of the computing power nodes, and the first target conversion coefficient, determine the target cost parameter.

[0026] In the above solution, the determining the target carbon emission parameter based on the first energy consumption parameter generated by the data transmission, the second energy consumption parameter generated by computing the sub-computation tasks on the computing power nodes, the current carbon emission parameter of the computing power nodes, the first target conversion coefficient, and the second target conversion coefficient includes:

[0027] Based on the first energy consumption parameter and the second energy consumption parameter, determine the target energy consumption parameter;

[0028] Based on the target energy consumption parameter, the first target conversion coefficient, and the second target conversion coefficient, determine the carbon emission parameter to be consumed;

[0029] Based on the current carbon emission parameter and the carbon emission parameter to be consumed, determine the target carbon emission parameter.

[0030] In the above solution, the determining the target cost parameter based on the first energy consumption parameter, the second energy consumption parameter, the current cost parameter of the computing power nodes, and the first target conversion coefficient includes:

[0031] Based on the first energy consumption parameter and the second energy consumption parameter, determine the target energy consumption parameter;

[0032] Based on the target energy consumption parameter and the first target conversion coefficient, determine the target power consumption, and based on the target power consumption, determine the cost parameter to be consumed;

[0033] Based on the current cost parameter and the cost parameter to be consumed, determine the target cost parameter.

[0034] In the above solution, determining the target computing power node for each group of sub-computation tasks from the sorted computing power nodes based on the first target weight, the second target weight, the third target weight, the target latency parameter, the target carbon emission parameter, and the target cost parameter includes:

[0035] According to the order of the levels of the multiple groups of sub-computation tasks from high to low, based on the first target weight, the second target weight, the third target weight, the target latency parameter, the target carbon emission parameter, and the target cost parameter, determine the target computing power node for each group of sub-computation tasks from the sorted computing power nodes.

[0036] In the above solution, determining the target computing power node for each group of sub-computation tasks from the sorted computing power nodes based on the first target weight, the second target weight, the third target weight, the target latency parameter, the target carbon emission parameter, and the target cost parameter includes:

[0037] For each group of sub-computation tasks, based on the first target weight, the second target weight, the third target weight, the target latency parameter, the target carbon emission parameter, and the target cost parameter, determine the corresponding second computing power value for each group of sub-computation tasks;

[0038] Determine the target computing power value from multiple second computing power values;

[0039] Based on the target computing power value, determine the target computing power node from the sorted computing power nodes.

[0040] An information determination device, the device includes:

[0041] An acquisition unit, configured to acquire the initial computing power resource parameter, the initial carbon emission parameter, and the initial cost parameter of each computing power node, and sort multiple computing power nodes based on the initial computing power resource parameter, the initial carbon emission parameter, and the initial cost parameter;

[0042] A receiving unit, configured to receive a computing power request carried by a client and the task label of the computing task;

[0043] A processing unit, configured to determine a target computing power node from the sorted computing power nodes based on the computing power request, the target computing power resource parameter, the target carbon emission parameter, the target cost parameter of the computing power node corresponding to the computing task, and the task label;

[0044] A sending unit, configured to send the identifier of the target computing power node to the client, so that the client sends the computing task to the target computing power node based on the identifier of the target computing power node.

[0045] An information determination device, the device comprising: a processor, a memory, and a communication bus;

[0046] The communication bus is used to implement a communication connection between the processor and the memory;

[0047] The processor is used to execute an information determination program stored in the memory to implement the steps of the above information determination method.

[0048] A computer-readable storage medium, the storage medium storing one or more programs, the one or more programs being executable by one or more processors to implement the steps of the information determination method as described above.

[0049] The information determination method, device, equipment, and computer-readable storage medium provided by the embodiments of the present application first obtain the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters of each computing power node, and sort multiple computing power nodes based on the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters. Then, a computing power request carrying a computing task and a task label of the computing task sent by the client is received. Next, based on the computing power request, the target computing power resource parameters, target carbon emission parameters, target cost parameters, and task label of the computing power node corresponding to the computing task, a target computing power node is determined from the sorted computing power nodes. After that, the identifier of the target computing power node is sent to the client, so that the client sends the computing task to the target computing power node based on the identifier of the target computing power node. In this way, when performing computing power scheduling, not only the initial computing power resource parameters (i.e., computing power, network topology, and delay parameters) of each computing power node are considered, but also the carbon emission parameters and cost parameters of each computing power node are considered. That is, the initial computing power resource parameters, carbon emission parameters, and cost parameters of each computing power node are comprehensively considered for computing power scheduling, so that the determined target computing power node meets the requirements of energy conservation and emission reduction. And, when performing computing power scheduling, the task label of the computing task is also considered, that is, hierarchical scheduling of the computing task can be performed based on the task label, improving the accuracy of computing for different computing tasks. Description of the Drawings

[0050] Figure 1 It is a schematic flowchart of an information determination method provided by an embodiment of the present application;

[0051] Figure 2 It is a schematic flowchart of another information determination method provided by an embodiment of the present application;

[0052] Figure 3 It is a schematic diagram of executing a computing task in an information determination method provided by an embodiment of the present application;

[0053] Figure 4Schematic diagram of computing power scheduling in an information determination method provided by an embodiment of the present application;

[0054] Figure 5 Schematic structural diagram of an information determination device provided by an embodiment of the present application;

[0055] Figure 6 Schematic structural diagram of an information determination device provided by an embodiment of the present application. Detailed implementation manners

[0056] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application.

[0057] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0058] It should be noted that in the traditional solution, the resource scheduling of the computing power network regards the computing power resources of the central cloud and the edge cloud as a whole, arranges all the computing power resources according to their static computing resource capabilities and the states of dynamic computing resources, generates task requirements in the computing power network according to the characteristics of the user's computing tasks, and selects an optimal data center for the user through a computing power scheduling or recommendation scheme. The traditional solution specifically includes relying on a computing power parsing service for the arrangement and allocation of computing power resources, and relying on the comparison of the computing power network integration characteristics between the data center and the user's computing request to select an optimal data center. Among them, the specific process of relying on the computing power parsing service for the arrangement and allocation of computing power resources is as follows: The computing power resources access the network and send a registration request to the computing power parsing server; the computing power parsing server abstracts the computing power resources hierarchically to generate a computing power resource identifier, and registers and manages the computing power resources; when the user initiates a computing power task, through a certain computing power resource matching mechanism, search for a list of available computing power resources on the computing power parsing server; after the computing power task is pre-allocated, the computing power network dynamically schedules and distributes tasks and exchanges and forwards data in real time according to the actual needs of the user; among them, the specific process of relying on the comparison of the computing power network integration characteristics between the data center and the user's computing request to select an optimal data center is as follows: Obtain the computing power network integration characteristics of the user's computing task requirements, and predict the computing power network integration characteristics of each data center according to a pre-established data center recommendation model; compare the computing power network integration characteristics of each data center with the computing power network integration characteristics of the user's computing task requirements, and recommend an optimal data center for the current user.

[0059] However, the above two traditional solutions have the following disadvantages: (1) Multi-level cloud collaboration is not considered: In the traditional solutions, the computing power resources on the central side and the edge side are normalized, without considering the characteristics of the computing power resources of the central cloud and the edge cloud, and how to perform the parsing, registration update, computing power request, computing power task scheduling and execution of the computing power resources for the moving client under the architecture of edge-cloud collaboration; (2) The impact factors of various computing power scheduling algorithms are not considered: The traditional solutions only consider the impact of computing power capabilities, network topologies, and latency parameters on computing power scheduling; (3) Hierarchical scheduling for tasks with different priorities is not performed.

[0060] Based on this, an information determination method is provided in an embodiment of the present application. This method can be applied to an information determination device. Referring to Figure 1 as shown, the method includes the following steps:

[0061] Step 101, obtain the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters of each computing power node, and sort the multiple computing power nodes based on the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters.

[0062] In the embodiment of the present application, the computing power node may refer to the computing power center on the central side and the computing power center on the edge side; among them, the computing power center on the central side (i.e., the computing power resources on the central side) is composed of a centralized computing power center or multiple distributed regional computing power centers, and the characteristic of the computing power resources on the central side is to use centralized and ultra-large-scale computing power resources to support computing tasks with low requirements for the latency of calculation results (such as training tasks) and computing tasks that cannot be executed on the edge-side computing power due to special reasons (such as the user cannot obtain edge-side computing power resources that meet the requirements); the computing power center on the edge side (i.e., the computing power resources on the edge side) refers to multiple distributed edge-cloud computing power centers, which can provide computing power resources for the computing tasks of the client at a position close to the client side.

[0063] In the embodiment of the present application, the initial computing power resource parameters may include computing power capabilities, initial latency parameters, and network topologies; the computing power parsing module in each computing power node can parse the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters of the computing power node in real time. After that, the computing power parsing module sends a registration request to the information determination device, and sends the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters to the information determination device after successful registration (i.e., information synchronization), so that the information determination device obtains the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters of each computing power node; it should be noted that the computing power parsing module can also update the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters to the scheduling center module in real time, so that the information determination device can obtain the computing power resources in the latest state of each computing power node.

[0064] In the embodiments of the present application, after the information determination device obtains the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters of each computing power node, it can uniformly manage the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters of each computing power node, and can sort multiple computing power nodes based on the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters of each computing power node. In this way, it is possible to comprehensively consider the cooperation between the central computing center and the edge computing center (i.e., the central cloud and the edge cloud) and effectively utilize the computing network fusion resources to support distributed computing tasks.

[0065] Step 102: Receive a computing power request sent by the client, which carries a computing task and a task label of the computing task.

[0066] In the embodiments of the present application, the computing task may refer to a task that the client (i.e., the user) needs to calculate through the computing power node; the client can attach a task label to each computing task; the client can send the computing power request to the information determination device in real time; it should be noted that when the client makes a change to the initially sent computing task, the client can also send a computing power request for the changed computing task to the information determination device to update the computing task.

[0067] In a feasible implementation manner, the computing task may be a training task of an Artificial Intelligence (AI) model submitted by the user, a long-range active safety warning task in the user's remote driving, a traffic congestion task on the traffic road. Of course, the computing task may also be other tasks.

[0068] Step 103: Determine a target computing power node from the sorted computing power nodes based on the computing power request, the target computing power resource parameters, target carbon emission parameters, target cost parameters of the computing power node corresponding to the computing task, and the task label.

[0069] In the embodiments of the present application, the target computing power node may refer to the computing power node that executes the computing task carried in the computing power request sent by the client and can meet the energy conservation and emission reduction requirements; each computing task can be decomposed into multiple sub-computing tasks; in the case of multiple sub-computing tasks, multiple target computing power nodes need to be determined; after receiving the computing power request, the information determination device can first respond to the computing power request, then group the multiple sub-computing tasks based on the task label of each sub-computing task, and then determine the target computing power node of each level of sub-computing tasks from the sorted computing power nodes according to the level based on the target computing power resource parameters, target carbon emission parameters, and target cost parameters.

[0070] It should be noted that multiple sub-computation tasks can be distributed among multiple edge-side computing power centers, can be distributed among multiple central-side computing power centers, can be distributed in the same edge-side computing power center, or can be distributed in the same central-side computing power center.

[0071] Step 104: Send the identifier of the target computing power node to the client, so that the client can send the computation task to the target computing power node based on the identifier of the target computing power node.

[0072] In the embodiments of the present application, the identifier of the computing power node can uniquely identify the computing power node; after the client obtains the identifier of the pre-allocated computing power node (i.e., the target computing power node) from the scheduling center module, it can submit the computation task to the target computing power node for computation; in a feasible implementation manner, the identifier of the target computing power node can refer to the address of the target computing power node.

[0073] The information determination method provided by the embodiments of the present application, when performing computing power scheduling, not only considers the initial computing power resource parameters (i.e., computing power capacity, network topology, and delay parameters) of each computing power node, but also considers the carbon emission parameters and cost parameters of each computing power node, that is, comprehensively considers the initial computing power resource parameters, carbon emission parameters, and cost parameters of each computing power node to perform computing power scheduling, so that the determined target computing power node meets the requirements of energy conservation and emission reduction; moreover, when performing computing power scheduling, it also considers the task tags of the computation tasks, that is, it can perform hierarchical scheduling on the computation tasks based on the task tags, improving the accuracy of computing different computation tasks.

[0074] Based on the foregoing embodiments, the embodiments of the present application provide another information determination method. Refer to Figure 2 As shown, this method may include the following steps:

[0075] Step 201: The information determination device determines the first weight of the initial computing power resource parameters, the second weight of the initial carbon emission parameters, and the third weight of the initial cost parameters of multiple computing power nodes.

[0076] Among them, the first weight may refer to the weight of the initial computing power resource parameters, the second weight may refer to the weight of the initial carbon emission parameters, and the third weight may refer to the weight of the initial cost parameters; the first weight, the second weight, and the third weight are all determined based on historical experimental data.

[0077] In the embodiments of the present application, when the initial computing power resource parameters include initial computing power capacity, initial delay parameters, and initial network topology, the first weight of the initial computing power capacity, the first weight of the initial delay parameters, and the first weight of the initial network topology can be determined respectively; among them, the first weight of the initial computing power capacity can be represented by k 1 to represent, and the first weight of the initial delay parameters can be represented by k 2is represented by k, the first weight of the initial network topology can be 3 is represented by k, the second weight of the initial carbon emission parameter can be 4 is represented by k, the third weight of the initial cost parameter can be 5 represented as follows.

[0078] Step 202: For each computing power node, the information determination device determines the first computing power value of each computing power node based on the initial computing power resource parameter, the initial carbon emission parameter, the initial cost parameter, the first weight, the second weight, and the third weight of the computing power node.

[0079] Among them, the first computing power value is used to represent the priority level of the computing power node.

[0080] In the embodiments of the present application, when the initial computing power resource parameter includes the initial computing power, the initial delay parameter, and the initial network topology, the initial computing power can be represented by C, the initial delay parameter can be represented by L, the initial network topology can be represented by N, the initial carbon emission parameter can be represented by E, and the initial cost parameter can be represented by M.

[0081] In the embodiments of the present application, the operations between the initial computing power resource parameter and the first weight, the initial carbon emission parameter and the second weight, and the initial cost parameter and the third weight can be performed first, and then the summation operation is performed on the above operation results to obtain the first computing power value of each computing power node; specifically, the first computing power value of each computing power node can be calculated by the following formula (1):

[0082]

[0083] It should be noted that the higher the first computing power value, the higher the priority level; the higher the initial computing power, the larger the corresponding value of the computing power, and the higher the priority level of the computing power node; the smaller the value of the initial network topology with a simple initial network topology, the higher the priority level of the computing power node; the lower the initial delay parameter, the higher the priority level of the computing power node; the smaller the value of the initial carbon emission parameter, the higher the priority level of the computing power node; the smaller the value of the initial cost parameter, the higher the priority level of the computing power node.

[0084] Step 203: The information determination device sorts the multiple computing power nodes based on the first computing power value.

[0085] In the embodiments of the present application, by sorting the multiple computing power nodes through the first computing power value, the sorted computing power nodes can be obtained. In this way, the hierarchical classification and sorting of the computing power center can be performed globally for subsequent computing power scheduling.

[0086] Step 204, the information determination device receives a computing power request sent by the client, which carries a computing task and a task label of the computing task.

[0087] Step 205, the information determination device groups multiple sub-computing tasks based on the computing power request and the task label to obtain multiple groups of sub-computing tasks.

[0088] Among them, the computing task includes multiple sub-computing tasks; the levels of each group of sub-computing tasks are different.

[0089] In the embodiment of the present application, the computing power request can be responded to first, and then multiple sub-computing tasks obtained by decomposing each computing task based on the task label are grouped to obtain multiple groups of sub-computing tasks with different levels. In this way, by dividing the levels of multiple sub-computing tasks, the multiple sub-computing tasks can be scheduled hierarchically, and for sub-computing tasks with different levels, different scheduling algorithms can also be used, improving the calculation accuracy of different computing tasks.

[0090] It should be noted that step 205 can be implemented in the following manner:

[0091] Step 205A1, the information determination device determines the level of each sub-computing task based on each task label.

[0092] In the embodiment of the present application, the level of the sub-computing task can refer to the priority of executing the sub-computing task; each task label can be analyzed to determine the level of each sub-computing task based on the analysis result.

[0093] Step 205A2, the information determination device determines the sub-computing tasks with the same level as a group of sub-computing tasks.

[0094] In the embodiment of the present application, sub-computing tasks with the same level can be classified to obtain multiple groups of sub-computing tasks with different levels, and the levels in each group of sub-computing tasks are the same; specifically, three groups of sub-computing tasks can be determined, including: high-level sub-computing tasks, medium-level sub-computing tasks, and low-level sub-computing tasks. Specifically, multiple high-level sub-computing tasks can be determined as a group of high-level sub-computing tasks, multiple medium-level sub-computing tasks can be determined as a group of medium-level sub-computing tasks, and multiple low-level sub-computing tasks can be determined as a group of low-level sub-computing tasks.

[0095] In a feasible implementation, if the sub-computation task is a real-time computation task (i.e., when the client is in a stationary or moving state in the transportation field and needs to complete real-time computations of a series of scenario services during the moving process, such as over-the-horizon active safety warnings in user remote driving, traffic congestion computations on traffic roads, etc.), the task label of the real-time computation task can be high level; if the sub-computation task is a training task (such as the training task of the AI model submitted by the client), the task label of the training task can be low level; if the sub-computation task is a short-term prediction task, the task label of the short-term prediction task can be high level; if the sub-computation task is a long-term prediction task, the task label of the long-term prediction task can be medium level; if the sub-computation task is a task with a short deadline, the task label of this task can be high level; if the sub-computation task is a task with a long deadline, the task label of this task can be low level; then, the real-time computation task, the short-term prediction task, and the task with a short deadline can be determined as a group of high-level sub-computation tasks, the long-term prediction task can be determined as a group of medium-level sub-computation tasks, and the training task and the task with a long deadline can be determined as a group of low-level sub-computation tasks.

[0096] Step 206: The information determination device determines the first target weight of the target computing power resource parameter, the second target weight of the target carbon emission parameter, and the third target weight of the target cost parameter for each group of sub-computation tasks.

[0097] In the embodiment of the present application, the first target weight can refer to the weight of the target computing power resource parameter, and in the case where the target computing power resource parameter includes the target delay parameter, the first target weight can refer to the weight of the target delay parameter, the second target weight can refer to the weight of the target carbon emission parameter, the third target weight can refer to the weight of the target cost parameter, and the first target weight can be represented by W T to represent, the second target weight can be represented by W C to represent, the third target weight can be represented by W I to represent; the first target weight, the second target weight, and the third target weight can all be obtained through historical experiment statistics optimization, and the first target weight, the second target weight, and the third target weight corresponding to each group of sub-computation tasks are different. Specifically, in the case of three groups of sub-computation tasks with different levels, and in the order of the levels of each group of sub-computation tasks from high to low, the different first target weights, second target weights, and third target weights assigned to multiple groups of sub-computation tasks are Among them, the first target weight W of the group of high-level sub-computation tasks can be set 11 to be larger, and the third target weight W of the group of low-level sub-computation tasks 33The first target weight W of a set of sub-computation tasks of a relatively large, medium level 21 , the second target weight W 22 and the third target weight W 23 are evenly distributed.

[0098] Step 207: The information determination device determines the target latency parameter of the computing power node corresponding to each set of sub-computation tasks, the target carbon emission parameter of the computing power node corresponding to each set of sub-computation tasks, and the target cost parameter of the computing power node corresponding to each set of sub-computation tasks.

[0099] Among them, the target computing power resource parameter includes the target latency parameter.

[0100] In the embodiment of the present application, the target latency parameter of the sub-computation task on the computing power node can be determined first, and the target energy consumption parameter of calculating the sub-computation task on the computing power node can be determined, and then the target carbon emission parameter and the target cost parameter can be determined based on the target energy consumption parameter.

[0101] It should be noted that step 207 can be implemented in the following manner:

[0102] Step 207B1: For each set of sub-computation tasks, the information determination device determines the target latency parameter based on the transmission latency parameter of data transmission between the sub-computation task and the computing power node and the computing latency parameter of calculating the sub-computation task on the computing power node.

[0103] In the embodiment of the present application, the transmission latency parameter (T j ) of data transmission between the sub-computation task (j) and the computing power node (i) can be determined first based on the data transmission volume (z ij ) and the data transmission rate (v tij ) of data transmission, and the computing latency parameter (T j ) of calculating the sub-computation task on the computing power node can be determined based on the computing power required by the sub-computation task (f i ) and the computing power provided by the computing power node for the sub-computation task (g rij ). Then, the sum operation is performed on T tij and T rij to obtain the target latency parameter (T ij ); specifically, T tij can be calculated by the following formula (2), and T rij can be calculated by the following formula (3), and T ij can be calculated by the following formula (4):

[0104]

[0105]

[0106] T ij = T tij + T rij Formula (4)

[0107] It should be noted that after step 207B1, step 207B2 can be executed first and then step 207B3, step 207B3 can be executed first and then step 207B2, or step 207B2 and step 207B3 can be executed simultaneously. In the embodiments of the present application, only the case of executing step 207B2 first and then step 207B3 is shown;

[0108] Step 207B2: The information determination device determines the target carbon emission parameter based on the first energy consumption parameter generated by data transmission, the second energy consumption parameter generated by calculating the sub-computation task on the computing power node, the current carbon emission parameter of the computing power node, the first target conversion coefficient, and the second target conversion coefficient.

[0109] In the embodiments of the present application, the first energy consumption parameter and the second energy consumption parameter can be operated first, then the operation result, the first target conversion coefficient, and the second target conversion coefficient can be operated, and then the target carbon emission parameter can be determined based on the operation result and the current carbon emission parameter.

[0110] It should be noted that step 207B2 can be implemented in the following manner:

[0111] Step 207b1: The information determination device determines the target energy consumption parameter based on the first energy consumption parameter and the second energy consumption parameter.

[0112] In the embodiments of the present application, the first energy consumption parameter can be determined first based on T tij and the conversion coefficient (p i ) between the delay parameter and the energy consumption parameter, and the second energy consumption parameter can be determined based on f and g j and the coefficient (k) related to the chip architecture of the computing power node i After that, the sum operation is performed on and to obtain the target energy consumption parameter (E ). Specifically, it can be calculated by the following formula (5) ij and calculated by the following formula (6) and calculated by the following formula (7) to obtain E : ij :

[0113]

[0114]

[0115]

[0116] Among them, k can take 10 -26 。

[0117] Step 207b2: The information determination device determines the carbon emission parameter to be consumed based on the target energy consumption parameter, the first target conversion coefficient, and the second target conversion coefficient.

[0118] In the embodiment of the present application, after obtaining the target energy consumption parameter E through the above formula ij subsequently, a multiplication operation can be performed on the target energy consumption parameter E ij , the first target conversion coefficient (δ), and the second target conversion coefficient (α) to obtain the carbon emission parameter to be consumed (C ij ), as shown in the following formula (8):

[0119] C ij = α·δ·E ij Formula (8)

[0120] Among them, α can take 0.785.

[0121] Step 207b3: The information determination device determines the target carbon emission parameter based on the current carbon emission parameter and the carbon emission parameter to be consumed.

[0122] In the embodiment of the present application, after determining the carbon emission parameter to be consumed (C ij ), a summation operation can be performed on the current carbon emission parameter (C t ) of the computing power node and the carbon emission parameter to be consumed (C ij ) to obtain the target carbon emission parameter.

[0123] Step 207B3: The information determination device determines the target cost parameter based on the first energy consumption parameter, the second energy consumption parameter, the current cost parameter of the computing power node, and the first target conversion coefficient.

[0124] In the embodiment of the present application, the first energy consumption parameter and the second energy consumption parameter can be operated first, then the operation result and the first target conversion coefficient can be operated, and then the target cost parameter can be determined based on the operation result and the current cost parameter.

[0125] It should be noted that step 207B3 can be implemented in the following manner:

[0126] Step 207b4: The information determination device determines the target energy consumption parameter based on the first energy consumption parameter and the second energy consumption parameter.

[0127] In the embodiment of the present application, a summation operation can be performed on the first energy consumption parameter and the second energy consumption parameter to obtain the target energy consumption parameter.

[0128] Step 207b5: The information determination device determines the target power based on the target energy consumption parameter and the first target conversion coefficient, and determines the cost parameter to be consumed based on the target power.

[0129] In the embodiment of the present application, the target energy consumption parameter E ij and the first target conversion coefficient (δ) can be multiplied to obtain the target power, and then the target power and the electricity price (β) can be multiplied to obtain the cost parameter to be consumed (I ij ), as shown in the following formula (9):

[0130] I ij = β·δ·E ij Formula (9)

[0131] Among them, the value of β can be the electricity price at the location of the computing power node.

[0132] Step 207b6: The information determination device determines the target cost parameter based on the current cost parameter and the cost parameter to be consumed.

[0133] In the embodiment of the present application, after determining the cost parameter to be consumed (I ij ), the current cost parameter (I t ) and the cost parameter to be consumed (I ij ) can be summed to obtain the target cost parameter; where t refers to the time t.

[0134] Step 208: The information determination device determines the target computing power node for each group of sub-computation tasks from the sorted computing power nodes based on the first target weight, the second target weight, the third target weight, the target latency parameter, the target carbon emission parameter, and the target cost parameter.

[0135] In the embodiment of the present application, for each group of sub-computation tasks, a target function can be established based on the first target weight, the second target weight, the third target weight, the target latency parameter, the target carbon emission parameter, and the target cost parameter, and the constraint conditions of the target function can be determined. Then, the target computing power node for each group of sub-computation tasks is determined from the sorted computing power nodes based on the target function and the constraint conditions.

[0136] It should be noted that Step 208 can be implemented in the following manner:

[0137] Step 208C: The information determination device determines the target computing power node for each group of sub-computation tasks from the sorted computing power nodes based on the first target weight, the second target weight, the third target weight, the target latency parameter, the target carbon emission parameter, and the target cost parameter in the order from high to low of the levels of the multiple groups of sub-computation tasks.

[0138] In the embodiments of the present application, in the case where multiple groups of sub-computation tasks include a group of high-level sub-computation tasks, a group of medium-level sub-computation tasks, and a group of low-level sub-computation tasks, parameters {W 11 , W 12 , W 13} can be selected for the group of high-level sub-computation tasks first, and computing power dispatching can be preferentially performed (that is, target computing power nodes are determined from the sorted computing power nodes), then parameters {W 21 , W 22 , W 23} are selected for the group of medium-level sub-computation tasks and computing power dispatching is performed, and then parameters {W 31 , W 32 , W 33} are selected for the group of low-level sub-computation tasks and computing power dispatching is performed. In this way, hierarchical scheduling of different levels of a group of sub-computation tasks can be achieved.

[0139] It should be noted that step 208C can also be implemented in the following manner:

[0140] Step 208c1: For each group of sub-computation tasks, an information determination device determines a corresponding second computing power value for each group of sub-computation tasks based on a first target weight, a second target weight, a third target weight, a target latency parameter, a target carbon emission parameter, and a target cost parameter.

[0141] In the embodiments of the present application, a target function as shown in the following formula (10) and constraint conditions as shown in the following formula (11) and formula (12) can be determined based on the first target weight, the second target weight, and the third target weight, the target latency parameter, the target carbon emission parameter, and the target cost parameter corresponding to the sub-computation tasks. Then, through reasonable computing power scheduling (that is, determining target computing power nodes from the sorted computing power nodes) based on the following formula (10), the comprehensive index of the latency parameter, the carbon emission parameter, and the cost parameter can be minimized, and the optimal system service performance and cost loss can be achieved:

[0142]

[0143] A t +∑ j∈N (f j ·S ij )≤A i Formula (11)

[0144]

[0145] It should be noted that M can refer to a total of M computing power nodes, N can refer to a total of N sub-computation tasks, A t is the computing power currently occupied by the computing power node, Ai It is the maximum computing power that computing power node i can provide.

[0146] Step 208c2: The information determination device determines a target computing power capability value from multiple second computing power capability values.

[0147] In the embodiment of the present application, the second computing power capability value with the smallest value among multiple second computing power capability values can be determined as the target computing power capability value.

[0148] Step 208c3: The information determination device determines a target computing power node from the sorted computing power nodes based on the target computing power capability value.

[0149] In the embodiment of the present application, the computing power node corresponding to the target computing power capability value is determined as the target computing power node; in a feasible implementation manner, as shown in Figure 3 In the case where the computing task is an over-the-horizon active safety warning task in user remote driving, this computing task can be decomposed into sub-computing task 1, sub-computing task 2, sub-computing tasks 3-9, and sub-computing task 10, and sub-computing task 1 is allocated to be calculated on edge-side computing power center 1, sub-computing task 2 is allocated to be calculated on edge-side computing power center 2, sub-computing tasks 3-9 are allocated to be calculated on edge-side computing power center n, and sub-computing task 10 is allocated to be calculated on the central-side computing power center.

[0150] In other embodiments of the present application, as shown in Figure 4 The computing power analysis module of the central-side computing power center can analyze the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters of the central-side computing power center, and the computing power analysis module of the edge-side computing power center can analyze the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters of the edge-side computing power center. After the analysis is completed, each computing power analysis module can send a registration request to the scheduling center module in the information determination device for computing power registration. In the case of successful registration, each computing power analysis module sends the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters to the scheduling center module, and then each computing power analysis module can also update the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters. Then, the computing power resource management module in the scheduling center module manages the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters of each central-side computing power center and edge-side computing power center (i.e., computing power resource management). Then, when the scheduling center module receives a computing power request sent by the user, the computing power scheduling module in the scheduling center module performs computing power allocation (i.e., sending a linked list of target computing power nodes to the user). Then, the user sends a computing task to the target computing power node based on the address in the linked list (i.e., computing power acquisition).

[0151] It should be noted that for the descriptions of the same steps and the same content in this embodiment and other embodiments, reference may be made to the descriptions in other embodiments, and details are not repeated here.

[0152] When the information determination method provided in the embodiment of the present application performs computing power scheduling, it not only considers the initial computing power resource parameters (i.e., computing power capacity, network topology, and delay parameters) of each computing power node, but also considers the carbon emission parameters and cost parameters of each computing power node. That is, when performing computing power scheduling, it comprehensively considers the initial computing power resource parameters, carbon emission parameters, and cost parameters of each computing power node, so that the determined target computing power node meets the requirements of energy conservation and emission reduction. Moreover, when performing computing power scheduling, it also considers the task tags of the computing tasks, that is, it can perform hierarchical scheduling on the computing tasks based on the task tags, improving the accuracy of computing for different computing tasks.

[0153] Based on the foregoing embodiments, the embodiment of the present application provides an information determination device, which can be applied to Figure 1 and Figure 2 In the information determination method provided in the corresponding embodiment, as shown in Figure 5 The information determination device 3 may include: an acquisition unit 31, a reception unit 32, a processing unit 33, and a transmission unit 34, where:

[0154] The acquisition unit 31 is configured to acquire the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters of each computing power node, and sort multiple computing power nodes based on the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters;

[0155] The reception unit 32 is configured to receive a computing power request carrying a computing task and a task tag of the computing task sent by the client;

[0156] The processing unit 33 is configured to determine a target computing power node from the sorted computing power nodes based on the computing power request, the target computing power resource parameters, target carbon emission parameters, target cost parameters, and task tag of the computing power node corresponding to the computing task;

[0157] The transmission unit 34 is configured to send the identifier of the target computing power node to the client, so that the client sends the computing task to the target computing power node based on the identifier of the target computing power node.

[0158] In other embodiments of the present application, the acquisition unit 31 is further configured to perform the following steps:

[0159] Determine the first weight of the initial computing power resource parameters, the second weight of the initial carbon emission parameters, and the third weight of the initial cost parameters of multiple computing power nodes;

[0160] For each computing power node, based on the initial computing power resource parameters, initial carbon emission parameters, initial cost parameters, first weight, second weight, and third weight of the computing power node, determine the first computing power value of each computing power node; wherein, the first computing power value is used to characterize the priority of the computing power node.

[0161] Sort the multiple computing power nodes based on the first computing power value.

[0162] In other embodiments of the present application, the processing unit 33 is further configured to perform the following steps:

[0163] Group the multiple sub-computing tasks based on the computing power request and the task tag to obtain multiple groups of sub-computing tasks; wherein, the computing task includes multiple sub-computing tasks; the levels of each group of sub-computing tasks are different.

[0164] For each group of sub-computing tasks, determine the first target weight of the target computing power resource parameters, the second target weight of the target carbon emission parameters, and the third target weight of the target cost parameters.

[0165] Determine the target delay parameter of the computing power node corresponding to each group of sub-computing tasks, the target carbon emission parameter of the computing power node corresponding to each group of sub-computing tasks, and the target cost parameter of the computing power node corresponding to each group of sub-computing tasks; wherein, the target computing power resource parameters include the target delay parameter.

[0166] Based on the first target weight, second target weight, third target weight, target delay parameter, target carbon emission parameter, and target cost parameter, determine the target computing power node of each group of sub-computing tasks from the sorted computing power nodes.

[0167] In other embodiments of the present application, the processing unit 33 is further configured to perform the following steps:

[0168] Based on each task tag, determine the level of each sub-computing task.

[0169] Determine the sub-computing tasks with the same level as a group of sub-computing tasks.

[0170] In other embodiments of the present application, the processing unit 33 is further configured to perform the following steps:

[0171] For each group of sub-computing tasks, based on the transmission delay parameter of the data transmission between the sub-computing task and the computing power node and the computing delay parameter of computing the sub-computing task on the computing power node, determine the target delay parameter.

[0172] Based on the first energy consumption parameter generated by the data transmission, the second energy consumption parameter generated by computing the sub-computing task on the computing power node, the current carbon emission parameter of the computing power node, the first target conversion coefficient, and the second target conversion coefficient, determine the target carbon emission parameter.

[0173] Determine a target cost parameter based on a first energy consumption parameter, a second energy consumption parameter, a current cost parameter of a computing power node, and a first target conversion coefficient.

[0174] In other embodiments of the present application, the processing unit 33 is further configured to perform the following steps:

[0175] Determine a target energy consumption parameter based on the first energy consumption parameter and the second energy consumption parameter;

[0176] Determine a carbon emission parameter to be consumed based on the target energy consumption parameter, the first target conversion coefficient, and the second target conversion coefficient;

[0177] Determine a target carbon emission parameter based on the current carbon emission parameter and the carbon emission parameter to be consumed.

[0178] In other embodiments of the present application, the processing unit 33 is further configured to perform the following steps:

[0179] Determine a target energy consumption parameter based on the first energy consumption parameter and the second energy consumption parameter;

[0180] Determine a target power consumption based on the target energy consumption parameter and the first target conversion coefficient, and determine a cost parameter to be consumed based on the target power consumption;

[0181] Determine a target cost parameter based on the current cost parameter and the cost parameter to be consumed.

[0182] In other embodiments of the present application, the processing unit 33 is further configured to perform the following steps:

[0183] In the order from high to low of the levels of multiple groups of sub-computation tasks, determine a target computing power node for each group of sub-computation tasks from the sorted computing power nodes based on a first target weight, a second target weight, a third target weight, a target latency parameter, a target carbon emission parameter, and a target cost parameter.

[0184] In other embodiments of the present application, the processing unit 33 is further configured to perform the following steps:

[0185] For each group of sub-computation tasks, determine a second computing power value corresponding to each group of sub-computation tasks based on a first target weight, a second target weight, a third target weight, a target latency parameter, a target carbon emission parameter, and a target cost parameter;

[0186] Determine a target computing power value from multiple second computing power values;

[0187] Based on the target computing power value, determine a target computing power node from the sorted computing power nodes.

[0188] It should be noted that for the specific implementation process of the steps executed by each module in the embodiments of the present application, reference may be made toFigure 1 and Figure 2 The implementation process in the information determination method provided by the corresponding embodiment will not be elaborated here.

[0189] When the information determination device provided by the embodiment of the present application performs computing power scheduling, it not only considers the initial computing power resource parameters of each computing power node (i.e., computing power capacity, network topology, and delay parameters), but also considers the carbon emission parameters and cost parameters of each computing power node. That is, it comprehensively considers the initial computing power resource parameters, carbon emission parameters, and cost parameters of each computing power node to perform computing power scheduling, so that the determined target computing power node meets the requirements of energy conservation and emission reduction; moreover, when performing computing power scheduling, it also considers the task tags of the computing tasks, that is, it can perform hierarchical scheduling on the computing tasks based on the task tags, improving the accuracy of computing for different computing tasks.

[0190] Based on the foregoing embodiments, the embodiments of the present application provide an information determination device, which can be applied to Figure 1 and Figure 2 the information determination method provided by the corresponding embodiment, referring to Figure 6 as shown, the information determination device 4 may include: a processor 41, a memory 42, and a communication bus 43, where:

[0191] The communication bus 43 is used to implement the communication connection between the processor 41 and the memory 42;

[0192] The processor 41 is used to execute the information determination program in the memory 42 to implement the following steps:

[0193] Obtain the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters of each computing power node, and sort the multiple computing power nodes based on the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters;

[0194] Receive a computing power request sent by the client carrying the computing task and the task tag of the computing task;

[0195] Based on the computing power request, the target computing power resource parameters, target carbon emission parameters, target cost parameters, and task tag of the computing power node corresponding to the computing task, determine the target computing power node from the sorted computing power nodes;

[0196] Send the identifier of the target computing power node to the client, so that the client sends the computing task to the target computing power node based on the identifier of the target computing power node.

[0197] In other embodiments of the present application, the processor 41 is used to execute the information determination program in the memory 42 to sort the multiple computing power nodes based on the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters to implement the following steps:

[0198] Determine the first weight of the initial computing power resource parameters, the second weight of the initial carbon emission parameters, and the third weight of the initial cost parameters for multiple computing power nodes;

[0199] For each computing power node, based on the initial computing power resource parameters, initial carbon emission parameters, initial cost parameters, the first weight, the second weight, and the third weight of the computing power node, determine the first computing power value of each computing power node; wherein, the first computing power value is used to represent the priority of the computing power node;

[0200] Sort the multiple computing power nodes based on the first computing power value.

[0201] In other embodiments of the present application, the processor 41 is used to execute the information determination program in the memory 42 to determine the target computing power node from the sorted computing power nodes based on the computing power request, the target computing power resource parameters, target carbon emission parameters, target cost parameters, and task tags of the computing power node corresponding to the computing task, so as to implement the following steps:

[0202] Group the multiple sub-computing tasks based on the computing power request and the task tags to obtain multiple groups of sub-computing tasks; wherein, the computing task includes multiple sub-computing tasks; the levels of each group of sub-computing tasks are different;

[0203] For each group of sub-computing tasks, determine the first target weight of the target computing power resource parameters, the second target weight of the target carbon emission parameters, and the third target weight of the target cost parameters;

[0204] Determine the target delay parameter of the computing power node corresponding to each group of sub-computing tasks, the target carbon emission parameter of the computing power node corresponding to each group of sub-computing tasks, and the target cost parameter of the computing power node corresponding to each group of sub-computing tasks; wherein, the target computing power resource parameters include the target delay parameter;

[0205] Based on the first target weight, the second target weight, the third target weight, the target delay parameter, the target carbon emission parameter, and the target cost parameter, determine the target computing power node for each group of sub-computing tasks from the sorted computing power nodes.

[0206] In other embodiments of the present application, the processor 41 is used to execute the information determination program in the memory 42 to group the multiple sub-computing tasks based on the task tags to obtain multiple groups of sub-computing tasks, so as to implement the following steps:

[0207] Based on each task tag, determine the level of each sub-computing task;

[0208] Determine the sub-computing tasks with the same level as a group of sub-computing tasks.

[0209] In other embodiments of the present application, the processor 41 is configured to execute the information determination program in the memory 42 to determine the target latency parameter of the computing power node corresponding to each group of sub-computation tasks, the target carbon emission parameter of the computing power node corresponding to each group of sub-computation tasks, and the target cost parameter of the computing power node corresponding to each group of sub-computation tasks, so as to implement the following steps:

[0210] For each group of sub-computation tasks, based on the transmission latency parameter of data transmission between the sub-computation tasks and the computing power node and the computing latency parameter of computing the sub-computation tasks on the computing power node, determine the target latency parameter;

[0211] Based on the first energy consumption parameter generated by data transmission, the second energy consumption parameter generated by computing the sub-computation tasks on the computing power node, the current carbon emission parameter of the computing power node, the first target conversion coefficient, and the second target conversion coefficient, determine the target carbon emission parameter;

[0212] Based on the first energy consumption parameter, the second energy consumption parameter, the current cost parameter of the computing power node, and the first target conversion coefficient, determine the target cost parameter.

[0213] In other embodiments of the present application, the processor 41 is configured to execute the information determination program in the memory 42 to determine the target carbon emission parameter based on the first energy consumption parameter generated by data transmission, the second energy consumption parameter generated by computing the sub-computation tasks on the computing power node, the current carbon emission parameter of the computing power node, the first target conversion coefficient, and the second target conversion coefficient, so as to implement the following steps:

[0214] Based on the first energy consumption parameter and the second energy consumption parameter, determine the target energy consumption parameter;

[0215] Based on the target energy consumption parameter, the first target conversion coefficient, and the second target conversion coefficient, determine the carbon emission parameter to be consumed;

[0216] Based on the current carbon emission parameter and the carbon emission parameter to be consumed, determine the target carbon emission parameter.

[0217] In other embodiments of the present application, the processor 41 is configured to execute the information determination program in the memory 42 to determine the target cost parameter based on the first energy consumption parameter, the second energy consumption parameter, the current cost parameter of the computing power node, and the first target conversion coefficient, so as to implement the following steps:

[0218] Based on the first energy consumption parameter and the second energy consumption parameter, determine the target energy consumption parameter;

[0219] Based on the target energy consumption parameter and the first target conversion coefficient, determine the target power consumption, and based on the target power consumption, determine the cost parameter to be consumed;

[0220] Based on the current cost parameter and the cost parameter to be consumed, determine the target cost parameter.

[0221] In other embodiments of the present application, the processor 41 is configured to execute the information determination program in the memory 42 to determine, based on the first target weight, the second target weight, the third target weight, the target latency parameter, the target carbon emission parameter, and the target cost parameter, the target computing power node for each group of sub-computation tasks from the sorted computing power nodes, so as to implement the following steps:

[0222] According to the order from high to low of the levels of multiple groups of sub-computation tasks, determine the target computing power node for each group of sub-computation tasks from the sorted computing power nodes based on the first target weight, the second target weight, the third target weight, the target latency parameter, the target carbon emission parameter, and the target cost parameter.

[0223] In other embodiments of the present application, the processor 41 is configured to execute the information determination program in the memory 42 to determine, based on the first target weight, the second target weight, the third target weight, the target latency parameter, the target carbon emission parameter, and the target cost parameter, the target computing power node for each group of sub-computation tasks from the sorted computing power nodes, so as to implement the following steps:

[0224] For each group of sub-computation tasks, determine the second computing power value corresponding to each group of sub-computation tasks based on the first target weight, the second target weight, the third target weight, the target latency parameter, the target carbon emission parameter, and the target cost parameter;

[0225] Determine the target computing power value from multiple second computing power values;

[0226] Based on the target computing power value, determine the target computing power node from the sorted computing power nodes.

[0227] It should be noted that the specific description of the steps executed by the processor can refer to the implementation process in the information determination method provided in the corresponding embodiments of Figure 1 and Figure 2 which will not be elaborated here.

[0228] When performing computing power scheduling, the information determination device provided in the embodiments of the present application not only considers the initial computing power resource parameters (i.e., computing power, network topology, and latency parameters) of each computing power node, but also considers the carbon emission parameter and cost parameter of each computing power node, that is, comprehensively considers the initial computing power resource parameters, carbon emission parameter, and cost parameter of each computing power node to perform computing power scheduling, so that the determined target computing power node meets the requirements of energy conservation and emission reduction; moreover, when performing computing power scheduling, the task label of the computing task is also considered, that is, hierarchical scheduling of the computing task can be performed based on the task label, improving the computing accuracy for different computing tasks.

[0229] Based on the foregoing embodiments, an embodiment of the present application provides a computer-readable storage medium storing one or more programs, which can be executed by one or more processors to implement Figure 1 and Figure 2 the steps in the information determination method provided in the corresponding embodiment.

[0230] It should be noted that the above computer-readable storage medium may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a ferromagnetic random access memory (FRAM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM), etc.; it may also be various electronic devices including one or any combination of the above memories, such as a mobile phone, a computer, a tablet device, a personal digital assistant, etc.

[0231] It should be noted that in this article, the term "including", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including a..." does not exclude the existence of another identical element in the process, method, article or device including the element.

[0232] The serial numbers of the embodiments of the present application above are only for description and do not represent the advantages or disadvantages of the embodiments.

[0233] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-described example methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions for causing a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in various embodiments of the present application.

[0234] This application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products of the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, and the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate for implementing in the process Figure 1 one process or multiple processes and / or blocks Figure 1 a device for the function specified in one block or multiple blocks.

[0235] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing devices to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured product including an instruction device, and the instruction device implements in the process Figure 1 one process or multiple processes and / or blocks Figure 1 a device for the function specified in one block or multiple blocks.

[0236] These computer program instructions can also be loaded onto a computer or other programmable data processing devices, so that a series of operation steps are executed on the computer or other programmable devices to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable devices provide steps for implementing the function specified in Figure 1 one process or multiple processes and / or blocks Figure 1 a device for the function specified in one block or multiple blocks.

[0237] The above are only the preferred embodiments of the present application, and do not limit the patent scope of the present application. Any equivalent structural or equivalent process transformation made by using the content of the specification and drawings of the present application, or directly or indirectly applied to other related technical fields, shall be equally included in the patent protection scope of the present application.

Claims

1. An information determination method, characterized in that, the method includes: Obtain the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters of each computing power node, and sort multiple computing power nodes based on the initial computing power resource parameters, the initial carbon emission parameters, and the initial cost parameters; Receive a computing power request sent by a client, which carries a computing task and a task label of the computing task; Based on the computing power request, the target computing power resource parameters, target carbon emission parameters, target cost parameters of the computing power node corresponding to the computing task, and the task label, determine a target computing power node from the sorted computing power nodes; Send the identifier of the target computing power node to the client, so that the client sends the computing task to the target computing power node based on the identifier of the target computing power node.

2. The method according to claim 1, characterized in that, the sorting of multiple computing power nodes based on the initial computing power resource parameters, the initial carbon emission parameters, and the initial cost parameters includes: Determine a first weight of the initial computing power resource parameters, a second weight of the initial carbon emission parameters, and a third weight of the initial cost parameters of the multiple computing power nodes; For each computing power node, based on the initial computing power resource parameters, the initial carbon emission parameters, the initial cost parameters, the first weight, the second weight, and the third weight of the computing power node, determine a first computing power value of each computing power node; wherein, the first computing power value is used to represent the priority of the computing power node; Sort the multiple computing power nodes based on the first computing power value.

3. The method according to claim 1, characterized in that, the determination of the target computing power node from the sorted computing power nodes based on the computing power request, the target computing power resource parameters, the target carbon emission parameters, the target cost parameters of the computing power node corresponding to the computing task, and the task label includes: Group multiple sub-computing tasks based on the computing power request and the task label to obtain multiple groups of sub-computing tasks; wherein, the computing task includes the multiple sub-computing tasks; the levels of each group of sub-computing tasks are different; For each group of sub-computing tasks, determine a first target weight of the target computing power resource parameters, a second target weight of the target carbon emission parameters, and a third target weight of the target cost parameters; Determine the target delay parameter of the computing power node corresponding to each group of sub-computing tasks, the target carbon emission parameter of the computing power node corresponding to each group of sub-computing tasks, and the target cost parameter of the computing power node corresponding to each group of sub-computing tasks; wherein, the target computing power resource parameters include the target delay parameter; Based on the first target weight, the second target weight, the third target weight, the target delay parameter, the target carbon emission parameter, and the target cost parameter, determine the target computing power node of each group of sub-computing tasks from the sorted computing power nodes.

4. The method according to claim 3, characterized in that, Grouping multiple sub-computation tasks based on the task tags to obtain multiple groups of sub-computation tasks includes: Based on each of the task tags, determining the level of each sub-computation task; Determining the sub-computation tasks with the same level as a group of sub-computation tasks.

5. The method according to claim 3, wherein, Determining the target delay parameter of the computing power node corresponding to each group of sub-computation tasks, the target carbon emission parameter of the computing power node corresponding to each group of sub-computation tasks, and the target cost parameter of the computing power node corresponding to each group of sub-computation tasks includes: For each group of sub-computation tasks, based on the transmission delay parameter of data transmission between the sub-computation task and the computing power node and the computing delay parameter of computing the sub-computation task on the computing power node, determining the target delay parameter; Based on the first energy consumption parameter generated by the data transmission, the second energy consumption parameter generated by computing the sub-computation task on the computing power node, the current carbon emission parameter of the computing power node, the first target conversion coefficient, and the second target conversion coefficient, determining the target carbon emission parameter; Based on the first energy consumption parameter, the second energy consumption parameter, the current cost parameter of the computing power node, and the first target conversion coefficient, determining the target cost parameter.

6. The method according to claim 5, wherein, Determining the target carbon emission parameter based on the first energy consumption parameter generated by the data transmission, the second energy consumption parameter generated by computing the sub-computation task on the computing power node, the current carbon emission parameter of the computing power node, the first target conversion coefficient, and the second target conversion coefficient includes: Based on the first energy consumption parameter and the second energy consumption parameter, determining the target energy consumption parameter; Based on the target energy consumption parameter, the first target conversion coefficient, and the second target conversion coefficient, determining the carbon emission parameter to be consumed; Based on the current carbon emission parameter and the carbon emission parameter to be consumed, determining the target carbon emission parameter.

7. The method according to claim 5, wherein, Determining the target cost parameter based on the first energy consumption parameter, the second energy consumption parameter, the current cost parameter of the computing power node, and the first target conversion coefficient includes: Based on the first energy consumption parameter and the second energy consumption parameter, determining the target energy consumption parameter; Based on the target energy consumption parameter and the first target conversion coefficient, determining the target power consumption, and based on the target power consumption, determining the cost parameter to be consumed; Based on the current cost parameter and the cost parameter to be consumed, determining the target cost parameter.

8. The method according to claim 3, wherein, Determining the target computing power node for each group of sub-computation tasks from the sorted computing power nodes based on the first target weight, the second target weight, the third target weight, the target delay parameter, the target carbon emission parameter, and the target cost parameter includes: According to the order of the levels of the multiple groups of sub-computation tasks from high to low, based on the first target weight, the second target weight, the third target weight, the target latency parameter, the target carbon emission parameter, and the target cost parameter, determine the target computing power nodes for each group of sub-computation tasks from the sorted computing power nodes.

9. The method according to claim 8, wherein, the determining the target computing power nodes for each group of sub-computation tasks from the sorted computing power nodes based on the first target weight, the second target weight, the third target weight, the target latency parameter, the target carbon emission parameter, and the target cost parameter includes: For each group of sub-computation tasks, determine the second computing power value corresponding to each group of sub-computation tasks based on the first target weight, the second target weight, the third target weight, the target latency parameter, the target carbon emission parameter, and the target cost parameter; Determine the target computing power value from multiple second computing power values; Based on the target computing power value, determine the target computing power nodes from the sorted computing power nodes.

10. An information determination device, wherein, the device includes: an acquisition unit, configured to acquire the initial computing power resource parameters, initial carbon emission parameters, and initial cost parameters of each computing power node, and sort multiple computing power nodes based on the initial computing power resource parameters, the initial carbon emission parameters, and the initial cost parameters; a receiving unit, configured to receive a computing power request sent by a client, which carries a computing task and a task label of the computing task; a processing unit, configured to determine the target computing power nodes from the sorted computing power nodes based on the computing power request, the target computing power resource parameters, target carbon emission parameters, target cost parameters of the computing power nodes corresponding to the computing task, and the task label; a sending unit, configured to send the identifier of the target computing power node to the client, so that the client sends the computing task to the target computing power node based on the identifier of the target computing power node.

11. An information determination device, wherein, the device includes: a processor, a memory, and a communication bus; the communication bus is used to implement a communication connection between the processor and the memory; the processor is configured to execute an information determination program stored in the memory to implement the steps of the information determination method according to any one of claims 1 to 9.

12. A computer-readable storage medium, wherein, the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of the information determination method according to any one of claims 1 to 9.

Images